What to do about an old accepted, upvoted but plainly incorrect answer?

Question

This question has an answer that is not only upvoted, but marked as accepted and having claimed a bounty. The problem is that, unless I'm particularly stupid today (it happens), a large part of it is completely wrong, specifically:

Example: If you shoot the same photo, say a headshot, with a 200mm lens and, at the same distance, with a 35mm lens. Then take the image from the 35mm and crop it to match the image from the 200mm you will find the DOF/image identical.

The answer was provided by a user who appears to be long-innactive and unlikely to make any changes in response to comments. Being a somewhat uncertain newbie, would it be better to submit an edit for approval, or flag the answer with an explanation? I feel like there should be another opinion involved...

Answer 1

If you read the comments, you'll see an explanation that makes the answer a little more reasonable. Basically, as Matt had noted, you need to scale up the crop (or, for that matter, shrink the telephoto) so that they were the same size visually. Presuming sufficient detail from the sensor, the apparent depth of field will be similar.

Now, with respect to your question here...

1. The edit option is likely to be rejected. We generally refuse edits that change the meaning of an answer. In fact, I'd probably hand out a review ban to a user that approved it because then it wouldn't be the answer anymore.
2. The flag option is also likely to be rejected. One of the options I have as a moderator when rejecting a flag is that flags shouldn't be used for incorrect answers, that's what voting is for. It is extremely rare that a moderator would delete the answer on this basis.

If you feel that a different opinion is warranted, by all means supply it! Late answers enter a review queue here and all reviewers are encouraged to look for late gems and vote for them. It also brings fresh attention to an old question that could result in down votes on answers already there and bubble up better answers, assuming there are, to the top.

Answer 2

Provide another answer, carefully documenting your case (ideally, with all of: a simple explanation, a technical explanation, and empirical examples). If you're correct, it should rise to the top — and hopefully the original question-asker will come back and change the checkmark. (If that doesn't happen, we can figure something out.)

We just had another example of this at How can the aperture value written in Exif be larger than the nominal limit of the lens? — if you have enough site rep to see deleted answers, you can see that there was a perfectly-reasonable-sounding one with a lot of upvotes (and although you can't see this, which was previously accepted as the answer). But someone came along with the actual right answer, with a reference to authoritative documentation, and unlike some other Q&A sites (Yahoo Answers, I'm looking at you) we can actually fix this stuff.